The objective of this study was to identify factors associated with frequent needle exchange program (NEP) attendance by injection drug users (IDUs) in Vancouver, Canada. Data were examined from a case control study of recent HIV infection. IDUs with documented HIV seroconversion after January 1, 1994 (n = 89) and seronegative controls with two documented HIV-negative test results in the same period (n = 192) were asked about demographic and social information, drug injection and sexual behavior, and NEP attendance. Logistic regression was used to examine the effect of multiple variables on NEP attendance while adjusting for HIV status and other potential confounders. Frequent (> 1 time/week) versus nonfrequent (< or = 1 time/week) NEP attenders did not differ with respect to gender, age, ethnicity, education, or HIV serostatus. For men, multivariate analysis showed that frequent cocaine injection was the only variable independently associated with NEP attendance (adjusted odds ratio [AOR] = 3.9; 95% confidence interval [CI] = 1.8-8.3); for women, independently associated variables were frequency of any drug injection (AOR = 5.5; 95% CI = 1.7-17), shooting gallery attendance (AOR = 11.5; 95% CI = 2.2-66), and having a nonlegal source of income (AOR = 3.4; 95% CI = 1.0-12). Borrowing used needles was associated with frequent NEP attendance in the univariate analysis. The NEP in Vancouver attracts IDUs who are frequent injectors (especially men using cocaine) and who have high-risk behaviors or an unstable lifestyle. This finding reinforces the role of NEPs as potential focal points for intervention in this hard-to-reach population.